Text Classification And Application Research For MOOC Discussion Forums

In MOOC,forum interaction is not only the main form of communication for teachers and learners,but also an effective tool to stimulate group awareness and group participation.The data in the forum is mostly presented in the form of short texts,with concise content and general meanings.In the era of big data,with the emergence of a large number of electronic texts,more and more scholars have begun to pay attention to the analysis and mining of text data,studying how to use technical means to extract valuable information from massive texts for relevant decision makers and users.Provide corresponding support.Aiming at the problem of large amount of data and mixed information in the MOOC discussion forum,this article mainly does the following research:(1)Analyze the existing classification navigation structure and user types of the MOOC discussion area,and propose a forum post classification system based on different user needs: course content-related category C,course logistics-related category L,evaluation suggestion category E,and other categories N;Using different machine learning methods,combined with the proposed classification system,build 5text classification models to achieve automatic classification of forum posts,and compare and analyze the performance of different models.(2)The experimental results of each baseline model show that: for the post text data in the MOOC forum,the classification accuracy and stability of the deep learning algorithm are higher than that of the traditional machine learning algorithm,and the accuracy of a single neural network classification algorithm can reach 90% ? 93%.Based on the well-performing deep learning classification algorithm,this paper constructs a CNN-Bi LSTM classification model,which is more suitable for the education field curriculum review scene,and its classification accuracy rate can reach94.82%.(3)Using system development technology,this paper designs and implements a MOOC discussion forum visual analysis system based on the above-mentioned text classification research results.The system takes the proposed deep learning classification model as the core and realizes the forum data in the MOOC platform of Chinese universities.Course post acquisition,post text preprocessing,post automatic classification and visual analysis,etc.,provide useful information and technical support for users of various forums.